The size of data sets being collected and analyzed in the industry for business intelligence is growing rapidly, making traditional warehousing solutions prohibitively expensive. Hadoop a popular open-source map-reduce implementation which used companies like Yahoo, Facebook etc. to store process extremely large on commodity hardware. However, programming model very low level requires developers write custom programs are hard maintain reuse. In this paper, we present Hive, an solution built...

Traffic prediction is a core problem in the intelligent transportation system and has broad applications management planning, main challenge of this field how to efficiently explore spatial temporal information traffic data. Recently, various deep learning methods, such as convolution neural network (CNN), have shown promising performance prediction. However, it samples data regular grids input CNN, thus destroys structure road network. In paper, we introduce graph propose an optimized...

One of the challenging topics in Intelligent Transportation Systems (ITSs) is metro passenger flow prediction. It has great practical significance for daily crowd management and vehicle scheduling flow. Recently, Graph Convolutional Networks (GCN) represent a station metros by aggregating information stations directly indirectly connected with station, improve effectiveness predicting Despite its effectiveness, neighborhood aggregation scheme also brings two limitations First, it limits to...

The building height holds significant importance for comprehensively understanding urban morphology, enhancing planning, and fostering sustainable development. Although many methods using optical SAR images have been presented estimation, these fall short in capturing the influences of economic social attributes on height. In this study, we introduced a Nature-Economy-Society (NES) feature model to represent information, established multi-scale One-Dimensional (1-D) Convolutional Neural...

To address the challenges of solving many-objective flexible job-shop scheduling problem, this study proposes a loose non-dominated sorting genetic algorithm III (LNSGA-III), an enhancement (NSGA-III). First, dominance principle is proposed to overcome shortcomings low selection pressure and slow convergence under Pareto principle. Next, novel crossover operator without repair, named improved order crossover, presented fully preserve characteristics exchanged operations enhance exploration...

A hybrid model for predicting urban arterial travel time on the basis of so-called state-space neural networks (SSNNs) and extended Kalman filter (EKF) is presented. Previous research demonstrated that SSNNs can address complex nonlinear spatiotemporal problems. However, SSNN models require off-line training with large sets input–output data, presenting three main drawbacks: ( a) great amounts effort are involved in collecting, preparing, executing these sessions; b) as mapping changes over...

Currently, people gain easy access to an increasingly diverse range of mobile devices such as personal digital assistants (PDAs), smart phones, and handheld computers. As dynamic content has become dominant on the fast-growing World Wide Web (C. Yuan et al., 2003), it is necessary provide effective ways for users prevalent in a computing environment. During course browsing devices, requested first dynamically generated by remote server, then transmitted over wireless network, and, finally,...

In modern internet industries, deep learning based recommender systems have became an indispensable building block for a wide spectrum of applications, such as search engine, news feed, and short video clips. However, it remains challenging to carry the well-trained models online real-time inference serving, with respect time-varying web-scale traffics from billions users, in cost-effective manner. this work, we present JIZHI - Model-as-a-Service system that per second handles hundreds...

As the sharing economy develops and bike-sharing apps emerge, dockless system (DLBS) has become a competitive alternative to docked because of its convenience finding parking without physical docks. Meanwhile, new demands are rapidly increasing as DLBS expands, e.g., crowd-sourced re-balancing pre-ordering during rush hours. A more fine-grained destination prediction is required tackle these issues. In this paper, we propose probabilistic-trip-based method named P³M. To overcome...

Machine learning based traffic forecasting models leverage sophisticated spatiotemporal auto-correlations to provide accurate predictions of city-wide states. However, existing methods assume a reliable and unbiased environment, which is not always available in the wild. In this work, we investigate vulnerability propose practical adversarial attack framework. Specifically, instead simultaneously attacking all geo-distributed data sources, an iterative gradient-guided node saliency method...

Two distinct ways of using Kalman filters to address the problem short-term urban arterial travel time prediction have been presented in this paper. One is train a neural network by incorporating extended filter. This approach utilizes filter find optimal weight parameters networks. The other use Filter solve state space model which used describe dynamic changes transportation systems, and obtain accurate estimation traffic variables. former one can be treated as data-driven without more...

We consider the problem of learning a policy for Markov decision process consistent with data captured on state-action pairs followed by policy. parameterize using features associated pairs. The can be handcrafted or defined kernel functions in reproducing Hilbert space. In either case, set large and only small, unknown subset may need to used fit specific data. parameters such are recovered 1-regularized logistic regression. establish bounds difference between average reward estimated...

Summary Inclement weather, such as heavy rain, significantly affects road traffic flow operation, which may cause severe congestion in networks cities. This study investigates the effect of inclement rain events, on and proposes an integrated model for parameter forecasting during events. First, analysis historical observation data indicates that error volume has a significant linear correlation with mean precipitation, thus, accuracy can be considerably improved by applying this to correct...

Bid optimization for online advertising from single advertiser's perspective has been thoroughly investigated in both academic research and industrial practice. However, existing work typically assume competitors do not change their bids, i.e., the wining price is fixed, leading to poor performance of derived solution. Although a few studies use multi-agent reinforcement learning set up cooperative game, they still suffer following drawbacks: (1) They fail avoid collusion solutions where all...

Finding the k smallest/largest element of a large array, i.e., k-selection is fundamental supporting algorithm in data analysis. Due to fact that big born geo-distributed environments, it especially requires communication-efficient distributed k-selection, besides typical computation and memory efficiency. Moreover, sensitive organizations make privacy rigorous precondition for their participation such statistical analysis common profit. To this end, we propose Distributed Privacy-Aware...

Abstract To elucidate the mechanism of wood sandwich compression, response compressing yield stress to hygrothermal conditions was investigated in this study with respect preheating temperature (30–210 °C) and moisture content (MC, 0–100 %). An associated functional model developed predict based on measured MC wood. A 1 % increase or a 10 °C led decrease exceeding 0.1 MPa. Significant variations stress, 0.8 MPa, were observed between high layer(s) remaining along thickness when there an...

We consider the problem of learning a policy used by an agent in Markov decision process using state-action samples. focus on class parameterized policies and use ℓ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> -regularized logistic regression to train that best fits observed pairs (demonstrations). bound difference average reward trained original (regret) terms generalization error sensitivity parameters chain. Specifically, we...

We develop a method to learn bio-inspired motion control policy using data collected from hawkmoths navigating in virtual forest. A Markov Decision Process (MDP) framework is introduced model the dynamics of moths and sparse logistic regression used parameters data. The results show that do not favor detailed obstacle location information navigation, but rely heavily on optical flow. Using learned moth as starting point, we propose an actor-critic learning algorithm refine obtain can be by...

City traffic is a dynamic system of enormous complexity. Modeling and predicting city flow remains to be challenge task the main difficulties are how specify supply demands parameterize model. In this paper we attempt solve these problems with help large amount floating car data. We propose coarse-grained cellular automata model that simulates vehicles moving on uniform grids whose size much larger compared microscopic The car-car interaction in replaced by coupling between state variables...

The accuracy of the loop-based vehicle classification under various traffic conditions is greatly dependent upon capability clarifying phases or states flow. One challenge lies in identifying using variables that could be directly calculated from dual-loop data. In this paper, we present a hybrid method incorporates level service and K-means clustering methods for We apply "phase representative variables" to represent characteristics flow phase identification algorithm. By video vehicular...

With the improvement of hardware technology, novel scanners such as uEXPLORER obtain significantly higher image quality than conventional scanners, but they have yet to be widespread. In addition, due limitations scanning time and radiotracer dose, PET images obtained in primary hospitals usually contains a lot noise, which has an impact on quantitative medical analysis diagnostic accuracy. It is great interest use enhancement methods enable comparable scanners. Obviously, this task, we are...