Search for: All records

Learning to route has received significant research momentum as a new approach for the route planning problem in intelligent transportation systems. By exploring global knowledge of geographical areas and topological structures of road networks to facilitate route planning, in this work, we propose a novel Generative Adversarial Network (GAN) framework, namely Progressive Route Planning GAN (ProgRPGAN), for route planning in road networks. The novelty of ProgRPGAN lies in the following aspects: 1) we propose to plan a route with levels of increasing map resolution, starting on a low-resolution grid map, gradually refining it on higher-resolution grid maps, and eventually on the road network in order to progressively generate various realistic paths; 2) we propose to transfer parameters of the previous-level generator and discriminator to the subsequent generator and discriminator for parameter initialization in order to improve the efficiency and stability in model learning; and 3) we propose to pre-train embeddings of grid cells in grid maps and intersections in the road network by capturing the network topology and external factors to facilitate effective model learning. Empirical result shows that ProgRPGAN soundly outperforms the state-of-the-art learning to route methods, especially for long routes, by 9.46% to 13.02% in F1-measure on multiple large-scale real-world datasets. ProgRPGAN, moreover, effectively generates various realistic routes for the same query. 

Estimating the travel time for a given path is a fundamental problem in many urban transportation systems. However, prior works fail to well capture moving behaviors embedded in paths and thus do not estimate the travel time accurately. To fill in this gap, in this work, we propose a novel neural network framework, namely Deep Image-based Spatio-Temporal network (DeepIST), for travel time estimation of a given path. The novelty of DeepIST lies in the following aspects:1) we propose to plot a path as a sequence of -generalized images"which include sub-paths along with additional information, such as traffic conditions, road network and traffic signals, in order to harness the power of convolutional neural network model (CNN)on image processing; 2) we design a novel two-dimensional CNN, namely PathCNN, to extract spatial patterns for lines in images by regularization and adopting multiple pooling methods; and 3) we apply a one-dimensional CNN to capture temporal patterns among the spatial patterns along the paths for the estimation. Empirical results show that DeepIST soundly outperforms the state-of-the-art travel time estimation models by 24.37% to 25.64% of mean absolute error (MAE) in multiple large-scale real-world datasets. 

Citations of scientific papers and patents reveal the knowledge flow and usually serve as the metric for evaluating their novelty and impacts in the field. Citation Forecasting thus has various applications in the real world. Existing works on citation forecasting typically exploit the sequential properties of citation events, without exploring the citation network. In this paper, we propose to explore both the citation network and the related citation event sequences which provide valuable information for future citation forecasting. We propose a novel Citation Network and Event Sequence (CINES) Model to encode signals in the citation network and related citation event sequences into various types of embeddings for decoding to the arrivals of future citations. Moreover, we propose a temporal network attention and three alternative designs of bidirectional feature propagation to aggregate the retrospective and prospective aspects of publications in the citation network, coupled with the citation event sequence embeddings learned by a two-level attention mechanism for the citation forecasting. We evaluate our models and baselines on both a U.S. patent dataset and a DBLP dataset. Experimental results show that our models outperform the state-of-the-art methods, i.e., RMTPP, CYAN-RNN, Intensity-RNN, and PC-RNN, reducing the forecasting error by 37.76% - 75.32%. 

Road network is a basic component of intelligent transportation systems (ITS) in smart city. Informative representation of road networks is important as it is essential to a wide variety of ITS applications. In this paper, we propose a neural network representation learning model, namely Intersection of Road Network to Vector (IRN2Vec), to learn embeddings of road intersections that encode rich information in a road network by exploring geo-locality and intrinsic properties of intersections and moving behaviors of road users. In addition to model design, several issues unique to IRN2Vec, including data preparation for model training and various relationships among intersections, are examined. We evaluate the learned embeddings via extensive experiments on three real-world datasets using three downstream test cases, including prediction of traffic signals and crossings on intersections and travel time estimation. Experimental results show that the proposed IRN2Vec outperforms three existing methods, DeepWalk, LINE and Node2vec, in terms of F1-score in predicting traffic signals (22.21% to 23.84%) and crossings (8.65% to 11.65%), and mean absolute error (MAE) in travel time estimation (9.87% to 19.28%). 

Informative representation of road networks is essential to a wide variety of applications on intelligent transportation systems. In this article, we design a new learning framework, called Representation Learning for Road Networks (RLRN), which explores various intrinsic properties of road networks to learn embeddings of intersections and road segments in road networks. To implement the RLRN framework, we propose a new neural network model, namely Road Network to Vector (RN2Vec), to learn embeddings of intersections and road segments jointly by exploring geo-locality and homogeneity of them, topological structure of the road networks, and moving behaviors of road users. In addition to model design, issues involving data preparation for model training are examined. We evaluate the learned embeddings via extensive experiments on several real-world datasets using different downstream test cases, including node/edge classification and travel time estimation. Experimental results show that the proposed RN2Vec robustly outperforms existing methods, including (i) Feature-based methods : raw features and principal components analysis (PCA); (ii) Network embedding methods : DeepWalk, LINE, and Node2vec; and (iii) Features + Network structure-based methods : network embeddings and PCA, graph convolutional networks, and graph attention networks. RN2Vec significantly outperforms all of them in terms of F1-score in classifying traffic signals (11.96% to 16.86%) and crossings (11.36% to 16.67%) on intersections and in classifying avenue (10.56% to 15.43%) and street (11.54% to 16.07%) on road segments, as well as in terms of Mean Absolute Error in travel time estimation (17.01% to 23.58%). 

In this paper, we propose a novel representation learning framework, namely HIN2Vec, for heterogeneous information networks (HINs). The core of the proposed framework is a neural network model, also called HIN2Vec, designed to capture the rich semantics embedded in HINs by exploiting different types of relationships among nodes. Given a set of relationships specified in forms of meta-paths in an HIN, HIN2Vec carries out multiple prediction training tasks jointly based on a target set of relationships to learn latent vectors of nodes and meta-paths in the HIN. In addition to model design, several issues unique to HIN2Vec, including regularization of meta-path vectors, node type selection in negative sampling, and cycles in random walks, are examined. To validate our ideas, we learn latent vectors of nodes using four large-scale real HIN datasets, including Blogcatalog, Yelp, DBLP and U.S. Patents, and use them as features for multi-label node classification and link prediction applications on those networks. Empirical results show that HIN2Vec soundly outperforms the state-of-the-art representation learning models for network data, including DeepWalk, LINE, node2vec, PTE, HINE and ESim, by 6.6% to 23.8% ofmicro-f1 in multi-label node classification and 5% to 70.8% of MAP in link prediction. 

The Skip-gram with negative sampling (SGNS) method of Word2Vec is an unsupervised approach to map words in a text corpus to low dimensional real vectors. The learned vectors capture semantic relationships between co-occurring words and can be used as inputs to many natural language processing and machine learning tasks. There are several high-performance implementations of the Word2Vec SGNS method. In this paper, we introduce a new optimization called context combining to further boost SGNS performance on multicore systems. For processing the One Billion Word benchmark dataset on a 16-core platform, we show that our approach is 3.53x faster than the original multithreaded Word2Vec implementation and 1.28x faster than a recent parallel Word2Vec implementation. We also show that our accuracy on benchmark queries is comparable to state-of-the-art implementations.