The criticality of efficient traffic forecasting in Intelligent Transportation System (ITS) has garnered significant academic attention. This study addresses the prevalent issue of distribution shift in real-world datasets, which often degrades performance, and explores the effectiveness of the channel-independence (CI), a technique recently proposed to mitigate this issue. While Spatio-Temporal Graph Neural Networks (STGNNs) are noted for their flexibility to represent road structures, their designs typically lack the capability to integrate CI without disrupting the spatial relationships, potentially limiting the performance. We present a novel approach that successfully integrates CI into spatial-temporal forecasting by incorporating distinct temporal, spatial, and predefined graph structure information within each channel. Moreover, STGNNs frequently emphasize intricate designs, which result in increased computational demands while offering only marginal improvements in accuracy. This paper presents ST-MLP, a streamlined spatio-temporal model constructed exclusively from cascaded Multi-Layer Perceptron (MLP) modules and linear layers. Experimental results indicate that ST-MLP outperforms numerous existing STGNNs in both accuracy and computational efficiency. Our findings advocate for further investigation into more streamlined and effective neural network architectures within spatial-temporal forecasting research.

Traditional recurrent neural network architectures, such as long short-term memory neural networks (LSTM), have historically held a prominent role in time series forecasting (TSF) tasks. While the recently introduced sLSTM for Natural Language Processing (NLP) introduces exponential gating and memory mixing that are beneficial for long term sequential learning, its potential short memory issue is a barrier to applying sLSTM directly in TSF. To address this, we propose a simple yet efficient algorithm named P-sLSTM, which is built upon sLSTM by incorporating patching and channel independence. These modifications substantially enhance sLSTM's performance in TSF, achieving state-of-the-art results. Furthermore, we provide theoretical justifications for our design, and conduct extensive comparative and analytical experiments to fully validate the efficiency and superior performance of our model.

Accurate traffic flow prediction is crucial for intelligent transportation systems supporting emerging applications such as autonomous driving and vehicle-infrastructure cooperation. However, existing methods often struggle to effectively disentangle the inherent trend, seasonal, and noise components within traffic flow data, thereby limiting prediction accuracy. To address this issue, we propose MF-AttnBiLSTM, a novel hybrid framework combining signal processing and temporal attention-based deep learning model through a decompose-then-predict strategy. Our approach first employs moving average to extract the trend component and discrete Fourier transform to isolate dominant seasonal patterns from the residuals. Subsequently, a dual-stream architecture utilizes multi-head self-attention-enhanced bidirectional LSTMs to independently model the temporal dynamics of the decomposed trend and seasonal components. The final prediction aggregates the outputs from both streams. Extensive experiments on PeMS04 and PeMS07 datasets demonstrate that MF-AttnBiLSTM significantly outperforms state-of-the-art baselines and exhibits robustness across varying traffic conditions. Ablation studies further confirm the efficacy of each component, particularly highlighting the significant contribution of the signal decomposition stage to overall performance improvement.

Crime prediction (CP) plays a pivotal role in urban analytics, contributing significantly to personal safety and societal stability. Unlike conventional time series forecasting, CP faces unique difficulties due to the inherent sparsity of crime incidents, particularly within small spatial regions and limited time windows. This sparsity, coupled with the non-Gaussian distribution of crime data—characterized by an excess of zero events and over-dispersion—presents a critical challenge for the signal processing community. In this regard, we propose a novel framework, Spatial-Temporal Multivariate Zero-Inflated Negative Binomial Graph Neural Networks (STMGNN-ZINB), which integrates diffusion and convolutional graph networks to capture spatial, temporal, and multivariate dependencies. By leveraging a Zero-Inflated Negative Binomial distribution, the STMGNN-ZINB effectively models the over-dispersed and zero-heavy nature of crime data, significantly improving both prediction accuracy and confidence interval estimation. Experimental results on real-world datasets demonstrate that our STMGNN-ZINB outperforms state-of-the-art CP methods, offering a robust tool for crime early warning and explicable insights into urban crime dynamics.

Video Anomaly Detection (VAD) remains a fundamental yet formidable task in the video understanding community, with promising applications in areas such as information forensics and public safety protection. Due to the rarity and diversity of anomalies, existing methods only use easily collected regular events to model the inherent normality of normal spatial-temporal patterns in an unsupervised manner. Although such methods have made significant progress benefiting from the development of deep learning, they attempt to model the statistical dependency between observable videos and semantic labels, which is a crude description of normality and lacks a systematic exploration of its underlying causal relationships. Previous studies have shown that existing unsupervised VAD models are incapable of label-independent data offsets (e.g., scene changes) in real-world scenarios and may fail to respond to light anomalies due to the overgeneralization of deep neural networks. Inspired by causality learning, we argue that there exist causal factors that can adequately generalize the prototypical patterns of regular events and present significant deviations when anomalous instances occur. In this regard, we propose Causal Representation Consistency Learning (CRCL) to implicitly mine potential scene-robust causal variable in unsupervised video normality learning. Specifically, building on the structural causal models, we propose scene-debiasing learning and causality-inspired normality learning to strip away entangled scene bias in deep representations and learn causal video normality, respectively. Extensive experiments on benchmarks validate the superiority of our method over conventional deep representation learning. Moreover, ablation studies and extension validation show that the CRCL can cope with label-independent biases in multi-scene settings and maintain stable performance with only limited training data available.

This article investigates the influence of limited information on the efficacy of traffic signal control algorithms for supporting Intelligent Transportation Systems. Our project has collected several classic and trending intelligent traffic signal control schemes and evaluated algorithmic performance under constrained data conditions. In these simulating scenarios, access to certain traffic data is restricted. The findings reveal that although information scarcity generally degrades algorithm performance, algorithms that perform well with comprehensive data maintain their superiority even in data-limited environments. This underscores the necessity of designing resilient algorithms capable of adapting to varying levels of information availability, offering valuable insights for developing robust traffic control systems.

Intelligent Transportation Systems (ITS) have garnered considerable attention as a potential solution for addressing the conflict between the increasing demand for transportation and the constraints within transportation infrastructure. One pivotal facet of this field is the domain of traffic flow prediction. In this paper, we introduce an inventive methodology for traffic flow prediction, in which we employ CNN to capture the underlying traffic data trends, while Support Vector Regression (SVR) with the signature kernel is adapted to predict the residual components within the traffic data. We evaluated our approach through comprehensive experiments based on real world traffic data, and the results clearly demonstrate a significant improvement in prediction accuracy over both ablation models and alternative state-of-art baseline methods.

As a potential solution to the growing conflict between the increasing demand for transportation and the limited capacity of transportation infrastructure, Intelligent Transportation Systems have gained considerable attention for their effectiveness in improving the efficiency of existing transportation infrastructure and enhancing traffic safety. Among various research areas, traffic flow prediction is a vital application, and researchers have devoted a lot of effort to designing accurate and fast algorithms. Currently, to satisfy various performance requirements, hybrid prediction methods that can take advantage of different sub-modules are beginning to emerge and show advantages in prediction accuracy and timeliness over other prediction algorithms that rely solely on machine learning. In this paper, we introduce a novel high precise traffic flow prediction method by utilizing the Fourier analysis (FA)-assisted denoising. Briefly, three sub-modules are introduced. Singular Spectrum Analysis (SSA) module is able to filter the noise of the original data, FA module is applied to extract periodic features of the traffic flow, and Long Short-Term Neural Networks (LSTM) is utilized to predict the future trend of time series residuals. We conducted simulation experiments. The corresponding test results demonstrate a substantial improvement in the accuracy compared to pure sub-models and other machine learning methods.

How to effectively improve the traffic efficiency of the road network plays a crucial role in ensuring the regular operation of modern society. This is also a key concern in the field of intelligent transportation systems. As the basis for formulating traffic control strategies, efficient and accurate traffic flow forecasting is essential. Accordingly, various prediction methods have been proposed for addressing the traffic flow prediction issue. However, we notice that most researchers only take the accuracy performance as the primary evaluation criteria and do not consider the problem of time cost. Consequently, the timeliness of the prediction results cannot be guaranteed. In this case, no matter how high the accuracy of the prediction is, it cannot provide practical information for the formulation of traffic measures. Therefore, in this paper, by exploiting the dimension reduction ability of Auto-Encoder (AE), we proposed a time-efficient prediction method for a large-scale road network that significantly reduces the prediction processing time while ensuring prediction accuracy. We conducted simulation experiments, and the corresponding test results demonstrate a substantial improvement in the time efficiency of our method compared to the traditional methods.

How to effectively improve the efficiency of vehicle traffic in the road system will play an essential role in improving the operational efficiency of the traffic system while eliminating the energy consumption and environmental pollution problems caused in particular, and this is also a key concern in the field of intelligent transportation systems. Timely and accurate traffic flow prediction is regarded as the key to solve the above problems because it can effectively improve the efficiency of traffic flow management. Many prediction methods have been proposed and among them, Machine Learning (ML)-based forecasting methods have gradually become mainstream in recent years because of their inherent ability to learn and predict nonlinear features in traffic information. However, we notice that most of the existing ML-based traffic prediction methods were designed relying fully on historical data while ignoring the structure and the impacts of the whole road network. Therefore, in this paper, we proposed a mixed method to take both historical data and road networks into consideration. Based on the real-world dataset, we conducted simulation experiments. The corresponding test results demonstrate a substantial improvement in the prediction accuracy of our method compared to conventional ML-based methods.