No file available
To read the file of this research,
you can request a copy directly from the authors.
Distribution-aware Online Continual Learning for Urban Spatio-Temporal Forecasting
Preprints and early-stage research may not have been peer reviewed yet.
Abstract
Urban spatio-temporal (ST) forecasting is crucial for various urban applications such as intelligent scheduling and trip planning. Previous studies focus on modeling ST correlations among urban locations in offline settings, which often neglect the non-stationary nature of urban ST data, particularly, distribution shifts over time. This oversight can lead to degraded performance in real-world scenarios. In this paper, we first analyze the distribution shifts in urban ST data, and then introduce DOST, a novel online continual learning framework tailored for ST data characteristics. DOST employs an adaptive ST network equipped with a variable-independent adapter to address the unique distribution shifts at each urban location dynamically. Further, to accommodate the gradual nature of these shifts, we also develop an awake-hibernate learning strategy that intermittently fine-tunes the adapter during the online phase to reduce computational overhead. This strategy integrates a streaming memory update mechanism designed for urban ST sequential data, enabling effective network adaptation to new patterns while preventing catastrophic forgetting. Experimental results confirm DOST's superiority over state-of-the-art models on four real-world datasets, providing online forecasts within an average of 0.1 seconds and achieving a 12.89% reduction in forecast errors compared to baseline models.
ResearchGate has not been able to resolve any citations for this publication.
Spatio-Temporal Graph Neural Network (STGNN) has been used as a common workhorse for traffic forecasting. However, most of them require prohibitive quadratic computational complexity to capture long-range spatio-temporal dependencies, thus hindering their applications to long historical sequences on large-scale road networks in the real-world. To this end, in this paper, we propose BigST, a linear complexity spatio-temporal graph neural network, to efficiently exploit long-range spatio-temporal dependencies for large-scale traffic forecasting. Specifically, we first propose a scalable long sequence feature extractor to encode node-wise long-range inputs ( e.g. , thousands of time-steps in the past week) into low-dimensional representations encompassing rich temporal dynamics. The resulting representations can be pre-computed and hence significantly reduce the computational overhead for prediction. Then, we build a linearized global spatial convolution network to adaptively distill time-varying graph structures, which enables fast runtime message passing along spatial dimensions in linear complexity. We empirically evaluate our model on two large-scale real-world traffic datasets. Extensive experiments demonstrate that BigST can scale to road networks with up to one hundred thousand nodes, while significantly improving prediction accuracy and efficiency compared to state-of-the-art traffic forecasting models.
The prediction of traffic flow is a challenging yet crucial problem in spatial-temporal analysis, which has recently gained increasing interest. In addition to spatial-temporal correlations, the functionality of urban areas also plays a crucial role in traffic flow prediction. However, the exploration of regional functional attributes mainly focuses on adding additional topological structures, ignoring the influence of functional attributes on regional traffic patterns. Different from the existing works, we propose a novel module named POI-MetaBlock, which utilizes the functionality of each region (represented by Point of Interest distribution) as metadata to further mine different traffic characteristics in areas with different functions. Specifically, the proposed POI-MetaBlock employs a self-attention architecture and incorporates POI and time information to generate dynamic attention parameters for each region, which enables the model to fit different traffic patterns of various areas at different times. Furthermore, our lightweight POI-MetaBlock can be easily integrated into conventional traffic flow prediction models. Extensive experiments demonstrate that our module significantly improves the performance of traffic flow prediction and outperforms state-of-the-art methods that use metadata.
The COVID-19 pandemic created an enormous disruption to the everyday life of the modern society. Among the various urban systems, transportation services were among those that suffered the most significant impacts, particularly severe in the case of highways. This paper addresses the challenges and responses to the pandemic from a private highway operator’s perspective and from a multidisciplinary perspective. Highway operators faced two main challenges: on one hand, the need to cope with the potential disruption caused by the pandemic and a national lockdown for almost three months, the provision of road services, and the requirement to ensure the proper operation and maintenance, and on the other hand, the strong negative impact of the pandemic on levels of traffic. Our case study shows that the operator’s management response in question is essentially characterised by being a first response to short term impacts while balancing for workers health and safety, engineering and management, internal business management, and overall economic impact. Highway operators were hardly prepared for such an event and became more focused on prioritising their employees and clients’ safety to avoid service disruption. Regarding levels of traffic, the pandemic has had severe effects, although to a varying degree, depending on the different types of vehicles (heavy, light, passenger, freight, among other types of vehicles) and the location of highways (coastal vs. interior). The lessons learnt can be valuable in future disruptive events and for other highway concession operators.
Critical incident stages identification and reasonable prediction of traffic incident duration are essential in traffic incident management. In this paper, we propose a traffic incident duration prediction model that simultaneously predicts the impact of the traffic incidents and identifies the critical groups of temporal features via a multi-task learning framework. First, we formulate a sparsity optimization problem that extracts low-level temporal features based on traffic speed readings and then generalizes higher level features as phases of traffic incidents. Second, we propose novel constraints on feature similarity exploiting prior knowledge about the spatial connectivity of the road network to predict the incident duration. The proposed problem is challenging to solve due to the orthogonality constraints, non-convexity objective, and non-smoothness penalties. We develop an algorithm based on the alternating direction method of multipliers (ADMM) framework to solve the proposed formulation. Extensive experiments and comparisons to other models on real-world traffic data and traffic incident records justify the efficacy of our model.
Spatial-temporal graph modeling is an important task to analyze the spatial relations and temporal trends of components in a system. Existing approaches mostly capture the spatial dependency on a fixed graph structure, assuming that the underlying relation between entities is pre-determined. However, the explicit graph structure (relation) does not necessarily reflect the true dependency and genuine relation may be missing due to the incomplete connections in the data. Furthermore, existing methods are ineffective to capture the temporal trends as the RNNs or CNNs employed in these methods cannot capture long-range temporal sequences. To overcome these limitations, we propose in this paper a novel graph neural network architecture, {Graph WaveNet}, for spatial-temporal graph modeling. By developing a novel adaptive dependency matrix and learn it through node embedding, our model can precisely capture the hidden spatial dependency in the data. With a stacked dilated 1D convolution component whose receptive field grows exponentially as the number of layers increases, Graph WaveNet is able to handle very long sequences. These two components are integrated seamlessly in a unified framework and the whole framework is learned in an end-to-end manner. Experimental results on two public traffic network datasets, METR-LA and PEMS-BAY, demonstrate the superior performance of our algorithm.
One crucial task in intelligent transportation systems is estimating the duration of a potential trip given the origin location, destination location as well as the departure time. Most existing approaches for travel time estimation assume that the route of the trip is given, which does not hold in real-world applications since the route can be dynamically changed due to traffic conditions, user preferences, etc. As inferring the path from the origin and the destination can be time-consuming and nevertheless error-prone, it is desirable to perform origin-destination travel time estimation, which aims to predict the travel time without online route information. This problem is challenging mainly due to its limited amount of information available and the complicated spatiotemporal dependency. In this paper, we propose a MUlti-task Representation learning model for Arrival Time estimation (MURAT). This model produces meaningful representation that preserves various trip properties in the real-world and at the same time leverages the underlying road network and the spatiotemporal prior knowledge. Further-more, we propose a multi-task learning framework to utilize the path information of historical trips during the training phase which boosts the performance. Experimental results on two large-scale real-world datasets show that the proposed approach achieves clear improvements over state-of-the-art methods
Timely accurate traffic forecast is crucial for urban traffic control and guidance. Due to the high nonlinearity and complexity of traffic flow, traditional methods cannot satisfy the requirements of mid-and-long term prediction tasks and often neglect spatial and temporal dependencies. In this paper, we propose a novel deep learning framework, Spatio-Temporal Graph Convolutional Networks (STGCN), to tackle the time series prediction problem in traffic domain. Instead of applying regular convolutional and recurrent units, we formulate the problem on graphs and build the model with complete convolutional structures, which enable much faster training speed with fewer parameters. Experiments show that our model STGCN effectively captures comprehensive spatio-temporal correlations through modeling multi-scale traffic networks and consistently outperforms state-of-the-art baselines on various real-world traffic datasets.
Accurate short-term traffic flow prediction is necessary for the implementation of Dynamic Route Guidance as motorists need to know traffic conditions ahead. The accuracy of short-term traffic flow prediction depends on how prediction models handle traffic flow characteristics such as temporal correlation, overdispersion, and seasonal patterns. Several data mining methods have been proposed to model and forecast traffic flow for the support of congestion control strategies. However, these methods focus on some of the characteristics and ignore others. Some methods address the autocorrelation and ignore the overdispersion and vice versa. In this research, we propose a data mining method that can consider all characteristics by capturing the flow autocorrelation, trend, and seasonality and by handling the overdispersion. The proposed method adopts the Holt-Winters-Taylor (HWT) count data method. Data from Taipei city are used to evaluate the proposed method which outperforms other methods by achieving a lower root mean square error. Then the proposed method is used in a dynamic route guidance systems to enhance the efficiency of guidance.
This paper introduces two recurrent neural network structures called Simple Gated Unit (SGU) and Deep Simple Gated Unit (DSGU), which are general structures for learning long term dependencies. Compared to traditional Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU), both structures require fewer parameters and less computation time in sequence classification tasks. Unlike GRU and LSTM, which require more than one gates to control information flow in the network, SGU and DSGU only use one multiplicative gate to control the flow of information. We show that this difference can accelerate the learning speed in tasks that require long dependency information. We also show that DSGU is more numerically stable than SGU. In addition, we also propose a standard way of representing inner structure of RNN called RNN Conventional Graph (RCG), which helps analyzing the relationship between input units and hidden units of RNN.
Taxicab is an important component of urban transit system since it caters to a large amount of demand and covers a wide geographic area. In this paper, we understand the spatial variation of urban taxi ridership using large scale New York City (NYC) taxi data. The taxi ridership is analyzed by relating it to various spatially explicit socio-demographic and built-environment variables. The geographically weighted regression (GWR) is implemented to model the spatial heterogeneity of the taxi ridership and visualize the spatial distributions of parameter estimations. The results suggest that the GWR model outperforms the ordinary least square model in both goodness of model fit and explanatory accuracy. The urban form is revealed to have significant impact on urban taxi ridership and strong spatial variability for parameter estimations is observed. Medium income level is found to reduce the number of taxi trips at particular places and the accessibility to subways is positively associated with the taxi ridership. The results provide valuable insights for predicting taxi demand as a function of spatially explicit variables which may have implications on taxi pricing, taxi industry regulation and urban planning.
With recent advances in sensing technologies, a myriad of spatio-temporal data has been generated and recorded in smart cities. Forecasting the evolution patterns of spatio-temporal data is an important yet demanding aspect of urban computing, which can enhance intelligent management decisions in various fields, including transportation, environment, climate, public safety, healthcare, and others. Traditional statistical and deep learning methods struggle to capture complex correlations in urban spatio-temporal data. To this end, Spatio-Temporal Graph Neural Networks (STGNN) have been proposed, achieving great promise in recent years. STGNNs enable the extraction of complex spatio-temporal dependencies by integrating graph neural networks (GNNs) and various temporal learning methods. In this manuscript, we provide a comprehensive survey on recent progress on STGNN technologies for predictive learning in urban computing. Firstly, we provide a brief introduction to the construction methods of spatio-temporal graph data and the prevalent deep-learning architectures used in STGNNs. We then sort out the primary application domains and specific predictive learning tasks based on existing literature. Afterward, we scrutinize the design of STGNNs and their combination with some advanced technologies in recent years. Finally, we conclude the limitations of existing research and suggest potential directions for future work.
Stream Learning (SL) attempts to learn from a data stream efficiently. A data stream learning algorithm should adapt to input data distribution shifts without sacrificing accuracy. These distribution shifts are known as ”concept drifts” in the literature. SL provides many supervised, semi-supervised, and unsupervised methods for detecting and adjusting to concept drift. On the other hand, Continual Learning (CL) attempts to preserve previous knowledge while performing well on the current concept when confronted with concept drift. In Online Continual Learning (OCL), this learning happens online. This survey explores the intersection of those two online learning paradigms to find synergies. We identify this intersection as Online Streaming Continual Learning (OSCL). The study starts with a gentle introduction to SL and then explores CL. Next, it explores OSCL from SL and OCL perspectives to point out new research trends and give directions for future research.
Accurately predicting traffic flow on roads is crucial to address urban traffic congestion and save on travel time. However, this is a challenging task due to the strong spatial and temporal correlations of traffic data. Existing traffic flow prediction methods based on graph neural networks and recurrent neural networks often overlook the dynamic spatiotemporal dependencies between road nodes and excessively focus on the local spatiotemporal dependencies of traffic flow, thereby failing to effectively model global spatiotemporal dependencies. To overcome these challenges, this article proposes a new Spatio-temporal Causal Graph Attention Network (STCGAT). STCGAT utilizes a node embedding technique that enables the generation of spatial adjacency subgraphs on a per-time-step basis, without requiring any prior geographic information. This obviates the necessity for intricate modeling of constantly changing graph topologies. Additionally, STCGAT introduces a proficient causal temporal correlation module that encompasses node-adaptive learning, graph convolution, as well as local and global causal temporal convolution modules. This module effectively captures both local and global Spatio-temporal dependencies. The proposed STCGAT model is extensively evaluated on traffic datasets. The results show that it outperforms all baseline models consistently.
As a core technology of Intelligent Transportation System, traffic flow prediction has a wide range of applications. The fundamental challenge in traffic flow prediction is to effectively model the complex spatial-temporal dependencies in traffic data. Spatial-temporal Graph Neural Network (GNN) models have emerged as one of the most promising methods to solve this problem. However, GNN-based models have three major limitations for traffic prediction: i) Most methods model spatial dependencies in a static manner, which limits the ability to learn dynamic urban traffic patterns; ii) Most methods only consider short-range spatial information and are unable to capture long-range spatial dependencies; iii) These methods ignore the fact that the propagation of traffic conditions between locations has a time delay in traffic systems. To this end, we propose a novel Propagation Delay-aware dynamic long-range transFormer, namely PDFormer, for accurate traffic flow prediction. Specifically, we design a spatial self-attention module to capture the dynamic spatial dependencies. Then, two graph masking matrices are introduced to highlight spatial dependencies from short- and long-range views. Moreover, a traffic delay-aware feature transformation module is proposed to empower PDFormer with the capability of explicitly modeling the time delay of spatial information propagation. Extensive experimental results on six real-world public traffic datasets show that our method can not only achieve state-of-the-art performance but also exhibit competitive computational efficiency. Moreover, we visualize the learned spatial-temporal attention map to make our model highly interpretable.
We all depend on mobility, and vehicular transportation affects the daily lives of most of us. Thus, the ability to forecast the state of traffic in a road network is an important functionality and a challenging task. Traffic data is often obtained from sensors deployed in a road network. Recent proposals on spatial-temporal graph neural networks have achieved great progress at modeling complex spatial-temporal correlations in traffic data, by modeling traffic data as a diffusion process. However, intuitively, traffic data encompasses two different kinds of hidden time series signals, namely the diffusion signals and inherent signals. Unfortunately, nearly all previous works coarsely consider traffic signals entirely as the outcome of the diffusion, while neglecting the inherent signals, which impacts model performance negatively. To improve modeling performance, we propose a novel Decoupled Spatial-Temporal Framework (DSTF) that separates the diffusion and inherent traffic information in a data-driven manner, which encompasses a unique estimation gate and a residual decomposition mechanism. The separated signals can be handled subsequently by the diffusion and inherent modules separately. Further, we propose an instantiation of DSTF, Decoupled Dynamic Spatial-Temporal Graph Neural Network (D ² STGNN), that captures spatial-temporal correlations and also features a dynamic graph learning module that targets the learning of the dynamic characteristics of traffic networks. Extensive experiments with four real-world traffic datasets demonstrate that the framework is capable of advancing the state-of-the-art.
Urban traffic status (e.g., traffic speed and volume) is highly dynamic in nature, namely, varying across space and evolving over time. Thus, predicting such traffic dynamics is of great importance to urban development and transportation management. However, it is very challenging to solve this problem due to spatial-temporal dependencies and traffic uncertainties. In this article, we solve the traffic dynamics prediction problem from Bayesian meta-learning perspective and propose a novel continuous spatial-temporal meta-learner (cST-ML), which is trained on a distribution of traffic prediction tasks segmented by historical traffic data with the goal of learning a strategy that can be quickly adapted to related but unseen traffic prediction tasks. cST-ML tackles the traffic dynamics prediction challenges by advancing the Bayesian black-box meta-learning framework through the following new points: (1) cST-ML captures the dynamics of traffic prediction tasks using variational inference, and to better capture the temporal uncertainties within tasks, cST-ML performs as a rolling window within each task; (2) cST-ML has novel designs in architecture, where CNN and LSTM are embedded to capture the spatial-temporal dependencies between traffic status and traffic-related features; (3) novel training and testing algorithms for cST-ML are designed. We also conduct experiments on two real-world traffic datasets (taxi inflow and traffic speed) to evaluate our proposed cST-ML. The experimental results verify that cST-ML can significantly improve the urban traffic prediction performance and outperform all baseline models especially when obvious traffic dynamics and temporal uncertainties are presented.
The Spatio-Temporal Traffic Prediction (STTP) problem is a classical problem with plenty of prior research efforts that benefit from traditional statistical learning and recent deep learning approaches. While STTP can refer to many real-world problems, most existing studies focus on quite specific applications, such as the prediction of taxi demand, ridesharing order, traffic speed, and so on. This hinders the STTP research as the approaches designed for different applications are hardly comparable, and thus how an application-driven approach can be generalized to other scenarios is unclear. To fill in this gap, this paper makes three efforts: (i) we propose an analytic framework, called STAnalytic, to qualitatively investigate STTP approaches regarding their design considerations on various spatial and temporal factors, aiming to make different application-driven approaches comparable; (ii) we design a spatio-temporal meta-model, called STMeta, which can flexibly integrate generalizable temporal and spatial knowledge identified by STAnalytic, (iii) we build an extensively large-scale STTP benchmark platform including ten datasets with five scenarios to quantitatively measure the generalizability of STTP approaches. In particular, we implement STMeta with different deep learning techniques, and STMeta demonstrates better generalizability than state-of-the-art approaches by achieving lower prediction error on average across all the datasets.
Traffic prediction has drawn increasing attention for its ubiquitous real-life applications in traffic management, urban computing, public safety, and so on. Recently, the availability of massive trajectory data and the success of deep learning motivate a plethora of deep traffic prediction studies. However, the existing neural-network-based approaches tend to ignore the correlations between multiple types of moving objects located in the same spatio-temporal traffic area, which is suboptimal for traffic prediction analytics.
In this paper, we propose a multi-source deep traffic prediction framework over spatio-temporal trajectory data, termed as MDTP. The framework includes two phases: spatio-temporal feature modeling and multi-source bridging. We present an enhanced graph convolutional network (GCN) model combined with long short-term memory network (LSTM) to capture the spatial dependencies and temporal dynamics of traffic in the feature modeling phase. In the multi-source bridging phase, we propose two methods, Sum and Concat, to connect the learned features from different trajectory data sources. Extensive experiments on two real-life datasets show that MDTP i) has superior efficiency, compared with classical time-series methods, machine learning methods, and state-of-the-art neural-network-based approaches; ii) offers a significant performance improvement over the single-source traffic prediction approach; and iii) performs traffic predictions in seconds even on tens of millions of trajectory data. we develop MDTP ⁺ , a user-friendly interactive system to demonstrate traffic prediction analysis.
Artificial neural networks thrive in solving the classification problem for a particular rigid task, acquiring knowledge through generalized learning behaviour from a distinct training phase. The resulting network resembles a static entity of knowledge, with endeavours to extend this knowledge without targeting the original task resulting in a catastrophic forgetting. Continual learning shifts this paradigm towards networks that can continually accumulate knowledge over different tasks without the need to retrain from scratch. We focus on task incremental classification, where tasks arrive sequentially and are delineated by clear boundaries. Our main contributions concern 1) a taxonomy and extensive overview of the state-of-the-art, 2) a novel framework to continually determine the stability-plasticity trade-off of the continual learner, 3) a comprehensive experimental comparison of 11 state-of-the-art continual learning methods and 4 baselines. We empirically scrutinize method strengths and weaknesses on three benchmarks, considering Tiny Imagenet and large-scale unbalanced iNaturalist and a sequence of recognition datasets. We study the influence of model capacity, weight decay and dropout regularization, and the order in which the tasks are presented, and qualitatively compare methods in terms of required memory, computation time and storage.
The advances in Internet of Things (IoT) and increased availability of road sensors allow for fine-grained traffic forecasting, which is of particular importance towards building an intelligent transportation system. In the literature, recent efforts have applied various deep learning methods for traffic forecasting, e.g., leveraging graph convolutional networks (GCNs) for spatial dependency modeling, and utilizing recurrent neural networks (RNNs) for capturing temporal dynamics. However, most of the existing approaches assume that spatial correlations are static and temporal correlations have only sequential dependencies and do not consider temporal periodicity of traffic across multiple time steps. The real challenge lies in using the dynamic spatio-temporal correlations while also considering the influence of non-traffic related factors such as time-of-day and weekday-or-weekend in the learning architectures. We propose a novel framework entitled “Reinforced Spatial-Temporal Attention Graph neural networks (RSTAG)" for traffic prediction. Our method captures dynamic spatial correlations through diffusion network graphs, while temporal dependencies are represented through the sequence-to-sequence model with an attention mechanism. In addition, we utilize policy gradient to update model parameters while largely alleviating the exposure bias issue that exists in previous traffic prediction models. We conduct extensive experiments on two large-scale traffic datasets collected from the road sensor networks in Los Angles and Bay Area of California. The results demonstrate that our method significantly outperforms the state-of-the-art baselines.
In taxi dispatch systems, predicting citywide passenger pickup/dropoff demand is indispensable for developing effective taxi distribution and scheduling strategies to resolve the demand-service mismatch. Compared with predicting next-step only, predicting multiple steps is preferable since it can provide a long term view, thus preventing short-sighted strategies. However, multi-step citywide passenger demand prediction (MsCPDP) is challenging due to the complicated spatiotemporal correlations in the distribution of passenger demand and the lack of ground truth from pre-steps for the prediction of subsequent steps. In this paper, a deep-learning-based prediction model with spatiotemporal attention mechanism is proposed for MsCPDP. The model, called ST-Attn, follows the general encoder-decoder framework for modelling sequential data but adopts a multiple-output strategy without recurrent neural network units. The spatiotemporal attention mechanism learns to determine the focus on those parts of the city at certain periods that are more relevant to the passenger demand in the predicted region and time period. In addition, a pre-predicted result calculated by spatiotemporal kernel density estimation is fed to ST-Attn, which provides a reference for further accurate prediction. Experiments on three real-world datasets are carried out to verify ST-Attn’s performance, and the results show that ST-Attn outperforms the baselines in terms of MsCPDP.
Ridesourcing, or on-demand ridesharing, is quickly changing today's travel. Recently, research has linked socio-demographics to ridesourcing use. However, little of the research has focused on the impacts of built environment, an important factor to consider in understanding travel behavior. This study applied Geographically Weighted Poisson Regression (GWPR) and examined the spatial relationships between built environment and ridesourcing demand. We used 2016-2017 ridesourcing trip data from a transportation network company (TNC), RideAustin, in Austin, Texas. By capturing the spatial heterogeneity, the GWPRs considerably improve modeling fit compared to the global models. Modeling results present strong relationships between ridesourcing demand and built environment variables (i.e., density, land use, infrastructure, and transit accessibility). More importantly, the results demonstrate significant spatial variations of the effects of both built environment and socioeconomic variables and geographic trends from urban to suburban neighborhoods. Overall, these findings suggest that built environment factors have significant impacts on ridesourcing demand, and it is important to consider the spatial context. The study provides useful insights for understanding ridesourcing use as a function of built environment and have important implications for transportation planning, demand modeling, and urban governance.
Concept drift describes unforeseeable changes in the underlying distribution of streaming data over time. Concept drift research involves the development of methodologies and techniques for drift detection, understanding and adaptation. Data analysis has revealed that machine learning in a concept drift environment will result in poor learning results if the drift is not addressed. To help researchers identify which research topics are significant and how to apply related techniques in data analysis tasks, it is necessary that a high quality, instructive review of current research developments and trends in the concept drift field is conducted. In addition, due to the rapid development of concept drift in recent years, the methodologies of learning under concept drift have become noticeably systematic, unveiling a framework which has not been mentioned in literature. This paper reviews over 130 high quality publications in concept drift related research areas, analyzes up-to-date developments in methodologies and techniques, and establishes a framework of learning under concept drift including three main components: concept drift detection, concept drift understanding, and concept drift adaptation. This paper lists and discusses 10 popular synthetic datasets and 14 publicly available benchmark datasets used for evaluating the performance of learning algorithms aiming at handling concept drift. Also, concept drift related research directions are covered and discussed. By providing state-of-the-art knowledge, this survey will directly support researchers in their understanding of research developments in the field of learning under concept drift.
Understanding the temporal traffic load profile of cellular networks is extremely valuable to many network operation tasks in large mobile networks. Such knowledge is useful for network planning, improving network performance, designing better load balancing schemes, testing handoff algorithms, and proposing new charging mechanisms. This paper proposes a simple yet powerful method to model the temporal traffic profile by a large 3G/LTE cellular network dataset in a metropolitan area, consisting of 9 thousand base stations and 3.5 million subscribers. Specifically, using the spectrum-based analysis, we extract three major frequency components, which captures the weekly, daily, and hourly temporal patterns in the traffic load across base stations. By clustering the traffic utilizing the features extracted from spectrum-domain components, we find that urban scale cellular traffic can be classified into five groups, which maps to five types of geographic locations. Besides the comprehensive analysis, we apply this model to two applications: predicting the future traffic load, and designing a load based pricing scheme, where we demonstrate the usefulness of our model and analysis results.
Taxi-calling apps are gaining increasing popularity for their efficiency in dispatching idle taxis to passengers in need. To precisely balance the supply and the demand of taxis, online taxicab platforms need to predict the Unit Original Taxi Demand (UOTD), which refers to the number of taxi-calling requirements submitted per unit time (e.g., every hour) and per unit region (e.g., each POI). Predicting UOTD is non-trivial for large-scale industrial online taxicab platforms because both accuracy and flexibility are essential. Complex non-linear models such as GBRT and deep learning are generally accurate, yet require labor-intensive model redesign after scenario changes (e.g., extra constraints due to new regulations). To accurately predict UOTD while remaining flexible to scenario changes, we propose LinUOTD, a unified linear regression model with more than 200 million dimensions of features. The simple model structure eliminates the need of repeated model redesign, while the high-dimensional features contribute to accurate UOTD prediction. We further design a series of optimization techniques for efficient model training and updating. Evaluations on two large-scale datasets from an industrial online taxicab platform verify that LinUOTD outperforms popular non-linear models in accuracy. We envision our experiences to adopt simple linear models with high-dimensional features in UOTD prediction as a pilot study and can shed insights upon other industrial large-scale spatio-temporal prediction problems.