Search

Predicting traffic flow represents a critical undertaking within the domain of Intelligent Transportation Systems (ITS), given its pivotal role in optimizing traffic management strategies. Nevertheless, this endeavor is fraught with challenges, stemming from the inherent complexity and dynamic nature of traffic systems. Existing methods still suffer from the following three main limitations: (i) modeling the spatio-temporal correlations in a static way, restricting the ability to learn dynamic correlations; (ii) only capturing short-term local temporal features while neglecting long-term temporal patterns; (iii) cannot learn the latent relationships between traffic nodes at different time steps. Therefore, we propose a novel method named InOutformer based on In -Time-Window self-attention and Out -Time-Window self-attention to predict traffic flow. This method can effectively capture both short-term local temporal features and model long-term global temporal patterns, while also capturing dynamic spatial correlations between nodes. Specifically, we first propose a novel time window partitioning mechanism to replace the conventional sliding time window approach, in order to capture dynamic temporal features. Second, two specially designed temporal self-attention modules are used to learn short-term local temporal features and model long-term global temporal patterns, respectively. Then, a 2D-attention method is devised to capture the spatial correlations of nodes at different time steps. Moreover, by exploiting transition flow data, the detrimental effects due to spatial information propagation delays are effectively mitigated. Extensive experiments on four real-world traffic datasets demonstrate that the proposed model InOutformer outperforms the baselines, including several state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Choosing appropriate Web services to create new applications plays a significant role for service-based development. The general service recommendation approaches usually prefer to suggest popular and frequently used services. However, nowadays users are becoming increasingly interested in long tail services, which are novel but less concerned. Unfortunately, few of the existing service recommendation methods explicitly takes such long tail services into consideration, leading to the poor diversity and serendipity of recommendation results. Compared with the general service recommendation, the major challenges of recommending long tail services come from: (1) more sparse historical invocation records, and (2) lower quality of textual service description. To address the above challenges, in this paper, we put forward a method for Long tail Service Recommendation based on Cross-view and Contrastive learning (LSRCC). More specifically, we propose a text view that employs recursive neural network, multilayer perceptron and attention mechanism to learn the semantic representations of application and service in a more precise way. In addition, we present an invocation view that employs LightGCN (Light Graph Convolution Network) and local level contrastive learning to capture the implicit connections among applications, services and their tags. Finally, we introduce global level contrastive learning to further integrate the above two views for the long tail service recommendation. The extensive experiments on real-world datasets demonstrate that LSRCC outperforms the state-of-the-art baselines in terms of diversity and tail coverage, with the satisfactory precision and recall, in long tail service recommendation. • A long tail service recommendation method is proposed for application creation. • Contrastive learning is used globally and locally to capture implicit connection. • Double views are used to mitigate the problem of data sparsity and inferiority. • Extensive experiments on the real datasets show its superiority over baselines. [ABSTRACT FROM AUTHOR]

Next point-of-interest (POI) recommendation has emerged as an essential task in recommender systems with the rapid development of location-based social networks (LBSNs). It has a wide range of applications in smart cities for building personalized scenarios. Current research typically uses sequential relationships to mine user preferences; however, it fails to sufficiently explore the spatial dependence of check-ins and the multi-perspective information they contain. To this end, this study proposes a Multiple Active Region Aware Network (MARAN), a novel routine-aware model for the next POI recommendation that simultaneously captures the user's routine regularity and short-term preference changes from check-in records. The key to MARAN is its ability to decompose sophisticated user behavior into two parts. One is a stable routine part characterized by central-based graphs built from historical trajectories based on spatial aggregation. The other is an unstable preference part that obtains the user's recent changes from short-term trajectories. Moreover, a neighborhood-aware negative sampler based on adjacent areas was designed to alleviate spatial sparsity, that is, the imbalance between positive and negative samples during model training. Experiments on two real-world datasets demonstrated that MARAN outperformed state-of-the-art methods. • Next POI recommendation considering spatial aggregation. • User's routines are characterized by central-based graphs. • Users' recent preferences are obtained from short-term trajectories. • Dynamic negative sampling based on current position. [ABSTRACT FROM AUTHOR]