On Network Embedding for Machine Learning on Road Networks: A Case Study on the Danish Road Network

Network embedding assigns nodes in a network to low-dimensional representations and effectively preserves the network structure. Recently, a significant amount of progresses have been made toward this emerging network analysis paradigm. In this survey, we focus on categorizing and then reviewing the current development on network embedding methods, and point out its future research directions. We first summarize the motivation of network embedding. We discuss the classical graph embedding algorithms and their relationship with network embedding. Afterwards and primarily, we provide a comprehensive overview of a large number of network embedding methods in a systematic manner, covering the structure- and property-preserving network embedding methods, the network embedding methods with side information and the advanced information preserving network embedding methods. Moreover, several evaluation approaches for network embedding and some useful online resources, including the network data sets and softwares, are reviewed, too. Finally, we discuss the framework of exploiting these network embedding methods to build an effective system and point out some potential future directions.

This paper presents models for calculating travel delays in heterogeneous road networks in extreme conditions. The models are developed to facilitate the calculation of fuel delivery and restoration delays in wireless communications base stations and other system nodes requiring electric power. The travel delay is calculated by representing road networks as spatial graphs whose attributes are modified by an extreme event. Travel delays are random and are characterized by a triangular distribution whose parameters, namely minimum, mode and maximum, are calculated using the developed models. Each road segment or edge's travel time is a function of its length and the type of road. The type of road is characterized by its speed limit. The travel delays to each cell site is calculated assuming that the crew and fuel trucks originate on the boundary of the analyzed region. A nuclear attack is used as an example of extreme event to exemplify the models. Results indicate that type of road, path choice and subsequent congestion post extreme event can have significant impacts on the spatial distribution of delays.

Network embedding: Taxonomies, frameworks and applications

Recent research on community detection focuses on learning representations of nodes using different network embedding methods, and then feeding them as normal features to clustering algorithms. However, we find that though one may have good results by direct clustering based on such network embedding features, there is ample room for improvement. More seriously, in many real networks, some statisticallysignificant nodes which play pivotal roles are often divided into incorrect communities using network embedding methods. This is because while some distance measures are used to capture the spatial relationship between nodes by embedding, the nodes after mapping to feature vectors are essentially not coupled any more, losing important structural information. To address this problem, we propose a general Markov Random Field (MRF) framework to incorporate coupling in network embedding which allows better detecting network communities. By smartly utilizing properties of MRF, the new framework not only preserves the advantages of network embedding (e.g. low complexity, high parallelizability and applicability for traditional machine learning), but also alleviates its core drawback of inadequate representations of dependencies via making up the missing coupling relationships. Experiments on real networks show that our new approach improves the accuracy of existing embedding methods (e.g. Node2Vec, DeepWalk and MNMF), and corrects most wrongly-divided statistically-significant nodes, which makes network embedding essentially suitable for real community detection applications. The new approach also outperforms other state-of-the-art conventional community detection methods.

Recently, network embedding has received a large amount of attention in network analysis. Although some network embedding methods have been developed from different perspectives, on one hand, most of the existing methods only focus on leveraging the plain network structure, ignoring the abundant attribute information of nodes. On the other hand, for some methods integrating the attribute information, only the lower-order proximities (e.g., microscopic proximity structure) are taken into account, which may suffer if there exists the sparsity issue and the attribute information is noisy. To overcome this problem, the attribute information and mesoscopic community structure are utilized. In this article, we propose a novel network embedding method termed Attributed Network Embedding with Micro-Meso structure, which is capable of preserving both the attribute information and the structural information including the microscopic proximity structure and mesoscopic community structure. In particular, both the microscopic proximity structure and node attributes are factorized by Nonnegative Matrix Factorization (NMF), from which the low-dimensional node representations can be obtained. For the mesoscopic community structure, a community membership strength matrix is inferred by a generative model (i.e., BigCLAM) or modularity from the linkage structure, which is then factorized by NMF to obtain the low-dimensional node representations. The three components are jointly correlated by the low-dimensional node representations, from which two objective functions (i.e., ANEM_B and ANEM_M) can be defined. Two efficient alternating optimization schemes are proposed to solve the optimization problems. Extensive experiments have been conducted to confirm the superior performance of the proposed models over the state-of-the-art network embedding methods.

Network embedding enables to apply off-the-shelf machine learning methods to the nodes on the network. Leveraging the textual information associated with nodes into network embedding methods is advantageous. However, only a few works try to leverage textual information to network embeddings. Moreover, the structure of networks and associated texts could be dynamically changed over time, this property leads shifts of the vector representation of nodes and words. However, to the best of our knowledge, none of previous network embedding methods considers chronological changes of vector representations of nodes and words. In this paper, we propose a dynamic text attribute network embedding method, which embeds the nodes and the words in a cooperative manner and takes chronological changes of vector representations of nodes and words into account. Experimental results show that (1) vector representations of nodes of our method achieve higher accuracy than baseline methods in classification and clustering tasks; (2) The vector representations of nodes and words successfully capture semantic similarity; And (3) our method successfully capture the chronological change of the vector representations over time.

On road networks, the clustering of moving objects is important for traffic monitoring and routes recommendation. The existing schemes find out density route by considering the number of vehicles in a road segment. Since they don’t consider the features of each road segment such as width, length, and directions in a road network, the results are not correct in some real road networks. To overcome such problems, we propose a clustering method for congested routes discovering from the trajectories of moving objects on road networks. The proposed scheme can be divided into three steps. First, it divides each road network into segments with different width, length, and directions. Second, the congested road segments are detected through analyzing the trajectories of moving objects on the road network. The saturation degree of each road segment and the average moving speed of vehicles in a road segment are computed to detect the congested road segments. Finally, we compute the final congested routes by using a clustering scheme. The experimental results showed that the proposed scheme can efficiently discover the congested routes in different directions of the roads.

Discrete network embedding emerged recently as a new direction of network representation learning. Compared with traditional network embedding models, discrete network embedding aims to compress model size and accelerate model inference by learning a set of short binary codes for network vertices. However, existing discrete network embedding methods usually assume that the network structures (e.g., edge weights) are readily available. In real-world scenarios such as social networks, sometimes it is impossible to collect explicit network structure information and it usually needs to be inferred from implicit data such as information cascades in the networks. To address this issue, we present an end-to-end discrete network embedding model for latent networks DELN that can learn binary representations from underlying information cascades. The essential idea is to infer a latent Weisfeiler-Lehman proximity matrix that captures node dependence based on information cascades and then to factorize the latent Weisfiler-Lehman matrix under the binary node representation constraint. Since the learning problem is a mixed integer optimization problem, an efficient maximal likelihood estimation based cyclic coordinate descent (MLE-CCD) algorithm is used as the solution. Experiments on real-world datasets show that the proposed model outperforms the state-of-the-art network embedding methods.

Travel time estimation (TTE) is widely applied for ride dispatching, ride-hailing, and route navigation. There are many factors affecting the travel time of a driver on a given trajectory, including the distance, road type, driving habits, traffic congestion, etc. Existing works fail to model the complex relationships of these factors for TTE. To fill this gap, in this paper, we first analyze how these factors work together in determining the travel time. In particular, the travel time depends on the distance and driving speed on each road segment of the trajectory, where the driving speed depends on the driving habits and the environment, including the static factors like the road type (highway or byway) and speed limit and the dynamic factor like the time of the day and congestion. Among these factors, driving habits and traffic conditions (e.g., jam) are the most difficult ones to model. Second, we propose to learn the driving habits of each driver via meta-learning and estimate the conditions based on the current and historical traffic conditions (via recurrent neural networks) of this road and its connected road segments (via graph convolutional neural network). The experimental results on two real taxi trajectory datasets show that our approach outperforms three state-of-the-art methods significantly.

Spatio-temporal graph convolutional networks for road network inundation status prediction during urban flooding

Due to the rapid development of vehicular transportation and urbanization, traffic congestion has been increasing and becomes a serious problem in almost all major cities worldwide. Many instances of traffic congestion can be traced to their root causes, the so-called traffic bottlenecks, where relief of traffic congestion at bottlenecks can bring network-wide improvement. Therefore, it is important to identify the locations of bottlenecks and very often the most effective way to improve traffic flow and relieve traffic congestion is to improve traffic situations at bottlenecks. In this article, we first propose a novel definition of traffic bottleneck taking into account both the congestion level cost of a road segment itself and the contagion cost that the congestion may propagate to other road segments. Then, an algorithm is presented to identify congested road segments and construct congestion propagation graphs to model congestion propagation in urban road networks. Using the graphs, maximal spanning trees are constructed that allow an easy identification of the causal relationship between congestion at different road segments. Moreover, using Markov analysis to determine the probabilities of congestion propagation from one road segment to another road segment, we can calculate the aforementioned congestion cost and identify bottlenecks in the road network. Finally, simulation studies using SUMO confirm that traffic relief at the bottlenecks identified using the proposed technique can bring more effective network-wide improvement. Furthermore, when considering the impact of congestion propagation, the most congested road segments are not necessarily bottlenecks in the road network. The proposed approach can better capture the features of urban bottlenecks and lead to a more effective way to identify bottlenecks for traffic improvement. Experiments are further conducted using data collected from inductive loop detectors in Taipei road network and some road segments are identified as bottlenecks using the proposed method.

Compact network embedding for fast node classification

Network embedding which aims to embed a given network into a low-dimensional vector space has been proved effective in various network analysis and mining tasks such as node classification, link prediction and network visualization. The emerging network embedding methods have shifted of emphasis in utilizing mature deep learning models. The neural-network based network embedding has become a mainstream solution because of its high efficiency and capability of preserving the nonlinear characteristics of the network. In this paper, we propose Adversarial Network Embedding using Structural Similarity (ANESS), a novel, versatile, low-complexity GAN-based network embedding model which utilizes the inherent vertex-to-vertex structural similarity attribute of the network. ANESS learns robustness and effective vertex embeddings via a adversarial training procedure. Specifically, our method aims to exploit the strengths of generative adversarial networks in generating high-quality samples and utilize the structural similarity identity of vertexes to learn the latent representations of a network. Meanwhile, ANESS can dynamically update the strategy of generating samples during each training iteration. The extensive experiments have been conducted on the several benchmark network datasets, and empirical results demonstrate that ANESS significantly outperforms other state-of-the-art network embedding methods.

The traffic road network is abstracted as a physical network, the intersection is the node, and the road segment is the edge. A road network set G(V, E) consisting of a node set and a road segment set is established, where V is a node (road junction) set and E is a side (road segment) set. Since some nodes in the road network will have a significant impact on the operation of the entire road network, it is important to identify the energy levels of the urban road network nodes. This paper proposes the concept of energy level, and uses the energy level recognition algorithm based on h-index to calculate the energy level of the road network node. Quickly and accurately judge and predict the traffic status and service level of urban road network intersections, which will help to induce traffic flow on urban road network, thereby improving the energy level of the entire city road network.

Objective: The objective in this study, using data from crashed cars fitted with on-board crash pulse recorders, was to present differences in average crash severity, distribution of crash severity, and injury outcomes, based on an independent safety rating of roads, also taking road type and speed limit into consideration. Furthermore, the objective was to evaluate differences in injury risk, based on the distribution of crash severity. Methods: The investigation included both frontal two-vehicle crashes and single-vehicle crashes with known injury outcome. In total, 209 real-world crashes involving cars fitted with crash pulse recorders were included. For all crashes, average mean acceleration and change of velocity of the vehicle acceleration pulse were measured and calculated. All crash spots were classified according to an independent road safety rating program (European Road Assessment Programme Road Protection Score), where the safety quality of roads is rated in relation to posted speed limits. The crash severity and injury outcome in crashes that occurred on roads with good safety ratings were compared with crashes on roads with poor safety ratings. The data were also divided into subcategories according to posted speed limit and road type, to evaluate whether there was a difference in crash severity and injury outcome within the categories. Results and Conclusions: In total, crash severity was statistically significantly lower in crashes occurring on roads with good safety ratings than in crashes occurring on roads with poor safety ratings. It was found that crash severity and injury risk were lower on roads with good safety ratings with a speed limit of above 90 km/h compared with roads with poor safety ratings, irrespective of speed limit. On the other hand, crash severity was higher on roads with good safety ratings with speed limit of 70 km/h than on roads with poor safety ratings with the same speed limit. Though it was found that a higher speed limit resulted in higher crash severity on roads with poor safety ratings, the opposite was found on roads with good safety ratings. The main reason for this was that lanes for traffic traveling in opposite directions were more often separated at higher speeds on roads with good safety ratings. On divided roads with good safety ratings, there were no crashes resulting in crash severity above the level corresponding to a 10 percent risk of sustaining serious or fatal injury. This indicates that one of the most important safety measures is divided roads.