Graph contrastive learning (GCL) has gained increasing interest as a solution for graph representation learning. In GCL, graph augmentation is essential to generate contrastive samples used for contrastive learning. Recently, most existing methods employ learnable graph view generators to augment graphs based on the node probability distribution adaptively. However, these methods cannot ensure that semantic-related nodes are preserved during graph augmentation, leading to performance degradation. To tackle this issue, we propose a novel approach called Semantic-aware Graph Contrastive Learning (SGCL), which can generate high-quality contrastive samples by only augmenting semantic-unrelated nodes so as to facilitate the performance of GCL on downstream tasks. Specifically, we first design a Lipschitz constant generator to compute the Lipschitz constants that measure the semantic relevance of each node. Then, we propose the Lipschitz graph augmentation to augment graphs while only dropping these semantic-unrelated nodes with small Lipschitz constants. Furthermore, we propose semanticaware contrastive learning to obtain more refined representations by contrasting the graph-level representation of anchor graphs and high-quality generated samples. Experimental results on unsupervised learning and transfer learning demonstrate the effectiveness of SGCL compared to state-of-the-art methods.

Accurate forecasting of traffic flow plays a crucial role in building smart cities in the new era. Previous work has achieved success in learning inherent spatial and temporal patterns of traffic flow. However, existing works investigated the multiple periodicities (e.g., hourly, daily, and weekly) of traffic via entanglement learning, which has not yet dealt with distribution shift and interaction shift problems in traffic flow. In this paper, we propose a novel disentanglement learning network, called MUSE-Net, to tackle the limitations of entanglement learning by simultaneously factorizing the exclusiveness and interaction of multi-periodic patterns in traffic flow. Grounded in the theory of mutual information, we first learn and dis-entangle exclusive and interactive representations of traffics from multi-periodic patterns. Then, we utilize semantic-pushing and semantic-pulling regularizations to encourage the learned representations to be independent and informative. Moreover, we derive a lower bound estimator to tractably optimize the disentanglement problem with multiple variables and propose a joint training model for traffic forecasting. Extensive experimental results on several real-world traffic datasets demonstrate the effectiveness of the proposed framework. The code is available at: https://github.com/JianyangQin/MUSE-Net.

Fake news detection attracts many researchers’ attention due to the negative impacts on the society. Most existing fake news detection approaches mainly focus on semantic analysis of news’ contents. However, the detection performance will dramatically decrease when the content of news is short. In this paper, we propose a novelfake news detection multi-task learning (FDML)model based on the following observations: 1) some certain topics have higher percentages of fake news; and 2) some certain news authors have higher intentions to publish fake news. FDML model investigates the impact of topic labels for the fake news and introduce contextual information of news at the same time to boost the detection performance on the short fake news. Specifically, the FDML model consists of representation learning and multi-task learning parts to train the fake news detection task and the news topic classification task, simultaneously. As far as we know, this is the first fake news detection work that integrates the above two tasks. The experiment results show that the FDML model outperforms state-of-the-art methods on real-world fake news dataset.

Taxi fraud has become a serious problem in many large cities, where passengers are overcharged by taxi drivers in various ways. Researchers have developed a number of methods to detect taxi frauds with the assumption that fraudulent trips, among normal trips, are recorded by taximeters. In this paper, different from the previous work, we identify a new type of taxi fraud called unmetered taxi rides, where taxi drivers carry passengers without activating the taximeters. Since these fraudulent rides are not recorded by taximeters, previous detection approaches cannot directly apply to them. Hence, we propose a novel fraud detection system specifically designed for unmetered taxi rides. Our system uses a learning model to detect unmetered trajectory segments that are similar to metered rides, and introduces a heuristic algorithm to construct maximum fraudulent trajectories from the trajectory dataset. We have conducted detailed experiments on real-world datasets, and the results show that the proposed system can detect unmetered taxi rides effectively and efficiently.

Big trajectory data introduces severe challenges for data storage and communication. In this paper, we propose a novel compression framework called Clockwise Compression Framework (CCF) for big trajectory data compression under road network constraints. In CCF, we design several new methods: 1) a spatial compression algorithm called Enhanced Clockwise Encoding (ECE), 2) a temporal compression algorithm called Fitting-based Temporal Simplification (FTS), and 3) a dedicated querier that processes queries based on the above spatial and temporal compression algorithms, without fully decompressing the trajectroy data. By leveraging the topological information of the road network, CCF is able to perform both spatial compression and temporal compression in on-line modes. We perform extensive experiments in a real big trajectory dataset to verify both effectiveness and efficiency of our methods. CCF shows promising performances in various metrics and outperforms the state-of-the-art methods.

The impact of inclement weather to urban traffic has been widely observed and studied for many years, with focus primarily on individual road segments by analyzing data from roadside deployed monitors. However, two fundamental questions are still open: (i) how to identify regional weather-traffic sensitivity index throughout a city, that indicates the degree to which the region traffic in a city is impacted by weather changes, (ii) among complex regional features, such as road structure and population density, how to dissect the most influential regional features that drive the urban region traffic to be more vulnerable to weather changes. Answering these questions is unprecedentedly important for urban planners to understand the functional characteristics of various urban regions throughout a city, and to improve traffic prediction and learn the key factors in urban planning. However, these two questions are nontrivial to answer, because urban traffic changes dynamically over time and is essentially affected by many other factors, which may dominate the overall impact. In this work, we make the first study on these questions, by developing a weather-traffic index (WTI) system. The system includes two main components: WTI establishment and key factor analysis. Using the proposed system, we conducted comprehensive empirical study in Shanghai, and the WTI extracted have been validated to be surprisingly consistent with real world observations. Further regional key factor analysis yields interesting results. For example, house age has significant impact on WTI, which sheds light on future urban planning and reconstruction.

Nowadays, there are many taxis traversing around the city searching for available passengers, but their hunts of passengers are not always efficient. To the dynamics of traffic and biased passenger distributions, current offline recommendations based on place of interests may not work well. In this paper, we define a new problem, global-optimal trajectory retrieving (GOTR), as finding a connected trajectory of high profit and high probability to pick up a passenger within a given time period in real-time. To tackle this challenging problem, we present a system, called HUNTS, based on the knowledge from both historical and online GPS data and business data. To achieve above objectives, first, we propose a dynamic scoring system to evaluate each road segment in different time periods by considering both picking-up rate and profit factors. Second, we introduce a novel method, called trajectory sewing, based on a heuristic method and the Skyline technique, to produce an approximate optimal trajectory in real-time. Our method produces a connected trajectory rather than several place of interests to avoid frequent next-hop queries. Third, to avoid congestion and other real-time traffic situations, we update the score of each road segment constantly via an online handler. Finally, we validate our system using a large-scale data of around 15,000 taxis in a large city in China, and compare the results with regular taxis' hunts and the state-of-the-art.

Nowadays, big cities are suffering from severe traffic congestion as a result of the continuing increase in vehicles. Taxis equipped with GPS can be viewed as sensors of the traffic situation in city. However, trajectory data generated by taxi's GPS traces are often high-dimensional and contain large spatial and temporal attributes, which pose challenges for analysts. In this paper, based on taxi trajectory data, we present an interactive visual analytics system, T-Watcher, for monitoring and analyzing complex traffic situations in big cities. Users are able to use a carefully designed interface to monitor and inspect data interactively from three levels (region, road and vehicle views). We develop a visualization method to monitor and analyze traffic patterns for abnormal behaviors detection. In the region view of our system, global temporal changes in spatial evolution will be presented to users and can be interactively explored. The road view shows temporal changes to the traffic situations of significant segments of roads. The vehicle view uses a novel visualization method to track individual vehicles. Furthermore, the three views integrate important statistical and historical information related to traffic, which illustrate temporal changes of the traffic. We find that this design can help users explore historical information while monitoring traffic. We test our system on a real-life vehicle dataset collected from thousands of taxis and obtained some interesting findings. The experimental results confirm the effectiveness and efficiency of the proposed visual detection method. The analysis of the results also shows that our system is capable of effectively monitoring traffic and detecting abnormal traffic patterns.