The advances in location-acquisition technologies have led to a myriad of spatial trajectories. These trajectories are usually generated at low or an irregular frequency due applications' characteristics energy saving, leaving the routes between two consecutive points single trajectory uncertain (called trajectory). In this paper, we present Route Inference framework based on Collective Knowledge (abbreviated as RICK) construct popular from Explicitly, given location sequence and time span,...

Dummy-based anonymization techniques for protecting location privacy of mobile users have been proposed in the literature. By generating dummies that move humanlike trajectories, shows can be preserved. However, by monitoring long-term movement patterns users, trajectories still exposed. We argue that, once trajectory a user is identified, locations Thus, it's critical to protect moving order preserve privacy. propose two schemes generate consistent long run. Guided three parameters...

In this paper, we try to improve the performance of particle swarm optimizer by incorporating linkage concept, which is an essential mechanism in genetic algorithms, and design a new identification technique called dynamic discovery address problem real-parameter optimization problems. Dynamic costless effective recognition that adapts configuration employing only selection operator without extra judging criteria irrelevant objective function. Moreover, recombination utilizes discovered...

The advance of object tracking technologies leads to huge volumes spatio-temporal data collected in the form trajectory stream. In this study, we investigate problem discovering groups that travel together (i.e., traveling companions) from Such technique has broad applications areas scientific transportation management and military surveillance. To discover companions, monitoring system should cluster objects each snapshot intersect clustering results retrieve moving-together objects. Since...

A Cyber-Physical System (CPS) integrates physical devices (e.g., sensors, cameras) with cyber (or informational)components to form a situation-integrated analytical system that responds intelligently dynamic changes of the real-world scenarios. One key issue in CPS research is trustworthiness analysis observed data: Due technology limitations and environmental influences, data are inherently noisy may trigger many false alarms. It highly desirable sift meaningful information from large...

The advance of mobile technologies leads to huge volumes spatio-temporal data collected in the form trajectory streams. In this study, we investigate problem discovering object groups that travel together (i.e., traveling companions ) from Such technique has broad applications areas scientific transportation management, and military surveillance. To discover companions, monitoring system should cluster objects each snapshot intersect clustering results retrieve moving-together objects. Since...

Sequential pattern mining is an important subfield in data mining. Recently, applications using time interval-based event have attracted considerable efforts discovering patterns from events that persist for some duration. Since the relationship between two intervals intrinsically complex, how to effectively and efficiently mine sequences a challenging issue. In this paper, novel representations, endpoint representation endtime representation, are proposed simplify processing of complex...

Network embedding has been proven effective to learn low-dimensional vector representations for network vertices, and recently received a tremendous amount of research attention. However, most existing methods merely focus on preserving the first second order proximities between nodes, important properties node centrality are neglected. Various measures such as Degree, Closeness, Betweenness, Eigenvector PageRank centralities have designed measure importance individual nodes. In this paper,...

In various web applications like targeted advertising and recommender systems, the available categorical features (e.g., product type) are often of great importance but sparse. As a widely adopted solution, models based on Factorization Machines (FMs) capable modelling high-order interactions among for effective sparse predictive analytics. volume web-scale data grows exponentially over time, analytics inevitably involves dynamic sequential features. However, existing FM-based assume no...

Ball trajectory data are one of the most fundamental and useful information in evaluation players' performance analysis game strategies. It is still challenging to recognize position a high-speed tiny ball accurately from an ordinary video. In this paper, we develop deep learning network, called TrackNet, track tennis broadcast videos which images small, blurry, sometimes with afterimage tracks or even invisible. The proposed heatmap-based network trained not only image single frame but also...

We demonstrate the first 7-core multicore erbium-doped fiber amplified (MC-EDFA) transmission of 40 x 128-Gbit/s PDM-QPSK signals over 6,160-km (MCF). The crosstalk (XT) from all other 6 cores a MC-EDFA and 55-km length MCF are about -46.5 dB -45.6 at center core, respectively. core-to-core rotation approach every span is used to average XT cores. averaged optical signal-to-noise ratio (OSNR) after 15.6 with 0.1 nm resolution bandwidth. Q-factor channels surpasses threshold...

With the popularity of social media (e.g., Facebook and Flicker), users can easily share their check-in records photos during trips. In view huge number user historical mobility in media, we aim to discover travel experiences facilitate trip planning. When planning a trip, always have specific preferences regarding Instead restricting limited query options such as locations, activities, or time periods, consider arbitrary text descriptions keywords about personalized requirements. Moreover,...

We study the forecasting problem for traffic with dynamic, possibly periodical, and joint spatial-temporal dependency between regions. Given aggregated inflow outflow of regions in a city from time slots 0 to <inline-formula><tex-math notation="LaTeX">$t - 1$</tex-math></inline-formula> , we predict at notation="LaTeX">$t$</tex-math></inline-formula> any region. Prior arts area often considered spatial temporal dependencies decoupled manner, or were rather computationally intensive training...

In this paper, we exploit the concept of sharing GPS data for estimate traffic information. Explicitly, implemented CarWeb, a platform to collect real-time from cars. Once receiving sufficient amount cars, are able information (i.e., speed roads). As such, users obtain update-to-date in CarWeb platform. addition, propose two algorithms with given. A prototype is and functionalities each component demonstrated paper.

Location prediction has attracted a significant amount of research effort. Given an object's recent movements and future time, the goal location is to predict this object at time specified. Prior works have elaborated on mining association relationships among regions, in which objects frequently appear, locations. Association regions are represented as rules. By exploring prior able good accuracy for prediction. However, with large trajectory data produced, huge rules expected. Furthermore,...

With the increasing number of mobile Apps developed, they are now closely integrated into daily life. In this paper, we develop a framework to predict that most likely be used regarding current device status smartphone. Such an usage prediction is crucial prerequisite for fast App launching, intelligent user experience, and power management smartphones. By analyzing real log data, discover two kinds features: The Explicit Feature (EF) from sensing readings built-in sensors, Implicit (IF)...

The increasing popularity and growth of mobile devices location-based services enable us to utilize large-scale geo-tagged data support novel applications. This paper introduces a problem called the best region search (BRS) provides efficient solutions it. Given set O spatial objects, submodular monotone aggregate score function, size x b query rectangle, BRS aims find rectangular such that objects inside is maximized. fundamental several real-world applications as most influential (eg....

Docked bike systems have been widely deployed in many cities around the world. To service provider, predicting demand and supply of bikes at any station is crucial to offering best quality. The docked prediction problem highly challenging because complicated joint spatial-temporal (ST) dependency as are picked up dropped off, so-called "flows", between stations. Prior works often considered spatial temporal dependencies separately using sequential network models, based on locality...

Most studies on sequential pattern mining are mainly focused time point-based event data. Few research efforts have elaborated patterns from interval-based However, in many real applications, usually persists for an interval of time. Since the relationships among intervals intrinsically complex, large database is really a challenging problem. In this paper, novel approach, named as incision strategy and new representation, called coincidence representation proposed to simplify processing...

The increasing demand for analyzing the insights in sports has stimulated a line of productive studies from variety perspectives, e.g., health state monitoring, outcome prediction. In this paper, we focus on objectively judging what and where to return strokes, which is still unexplored turn-based sports. By formulating stroke forecasting as sequence prediction task, existing works can tackle problem but fail model information based characteristics badminton. To address these limitations,...

Abstract Background Medical records are a valuable source for understanding patient health conditions. Doctors often use these to assess without solely depending on time-consuming and complex examinations. However, may not always be directly relevant patient’s current issue. For instance, information about common colds more specific condition. While experienced doctors can effectively navigate through unnecessary details in medical records, this excess presents challenge machine learning...