Article
Green & Sustainable Science & Technology
Sergio Nesmachnow, Renzo Massobrio, Santiago Guridi, Santiago Olmedo, Andrei Tchernykh
Summary: This article introduces a model based on big data analysis to evaluate the travel times of buses in public transportation systems. The results show a reasonably good level of punctuality in the public transportation system, with some variability in terms of speed.
Article
Green & Sustainable Science & Technology
Hui Zhang, Chengxiang Zhuge, Jianmin Jia, Baiying Shi, Wei Wang
Summary: This paper proposes a network-based method to detect the travel mobility of DBS users and analyzes the characteristics of DBS networks using a data-driven framework. The study reveals that travel demands of DBS are higher during morning and evening peaks on weekdays, with network connections concentrated in central areas.
JOURNAL OF CLEANER PRODUCTION
(2021)
Article
Automation & Control Systems
Guangqiang Xie, Runpeng Zhang, Yang Li, Ling Huang, Chang-Dong Wang, Hao Yang, Jiahao Liang
Summary: The article introduces a new district attraction ranking approach called AttractRank, which utilizes taxi big data for ranking and an application system development. The research demonstrates the effective use of urban data for urban planning, visualizing the attraction ranking of each district to discover interesting patterns about urban life.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Hospitality, Leisure, Sport & Tourism
Lina Zhong, Alastair M. Morrison, Chengjun Zheng, Xiaonan Li
Summary: This study aims to derive destination image attributes and establish a classification system for destinations based on online consumer narratives. Content and social network analyses were used to explore the consumer image structure, and cluster analysis and ANOVA were used to group destinations and compare them. The study identified 22 attributes and grouped destinations into three categories based on their network centralities. Landscape, traffic, food and beverages, and resource-based attractions were the most mentioned attributes. Social life was found to be meaningful in consumer narratives but often overlooked by researchers.
Article
Operations Research & Management Science
Claire Y. T. Chen, Edward W. Sun, Ming-Feng Chang, Yi-Bing Lin
Summary: With the increasing environmental concerns and the utilization of big data, smart transportation is transforming logistics business and operations towards a more sustainable approach. This paper presents a new deep learning approach called bi-directional isometric-gated recurrent unit (BDIGRU) for predictive analysis of travel time and business adoption for route planning. The proposed method directly learns high-level features from big traffic data and reconstructs them using its own attention mechanism, achieving significant improvements in predictive accuracy and efficient route determination under uncertainty.
ANNALS OF OPERATIONS RESEARCH
(2023)
Article
Computer Science, Artificial Intelligence
Huangke Chen, Xin Zhang, Ling Wang, Lining Xing, Witold Pedrycz
Summary: This study proposes a self-organized autonomous optimization approach for offloading large-scale resource-constrained satellites' Earth observation big data. The approach characterizes the relationship between resource availability and constraints through the gradient of each satellite and makes data offloading decisions using a neighborhood update strategy and a bidirectional selection-based optimization strategy. Empirical results demonstrate the significant advantage of this approach in shortening response time.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Chengshun Huang, Li Zhu
Summary: This paper investigates the evaluation method of communication network robustness based on big data and cloud edge computing. By analyzing the basic models of complex networks and conducting experiments, it reveals the robustness of different types of networks.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Transportation Science & Technology
Homayoun Hamedmoghadam, Hai L. Vu, Mahdi Jalili, Meead Saberi, Lewi Stone, Serge Hoogendoorn
Summary: The origin-destination travel demand matrix plays a crucial role in analyzing transportation systems, particularly in public transportation networks. This study proposes a statistical pattern recognition approach to effectively extract this matrix from passenger smartcard records, overcoming challenges like 'alighting transaction inference' and 'transfer identification'. The framework is tested on a large dataset from Melbourne's public transportation network, showing promising results in accurately estimating the demand matrix.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2021)
Article
Computer Science, Artificial Intelligence
Xiujuan Xu, Yuzhi Sun, Yulin Bai, Kai Zhang, Yu Liu, Xiaowei Zhao
Summary: This paper introduces a novel model based on convolutional neural network, which converts sequence data into image data and combines dense connection network and residual network to effectively solve the travel time index prediction problem of complex road networks at the regional level.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Interdisciplinary Applications
Qiangqiang Xiong, Yaolin Liu, Peng Xie, Yiheng Wang, Yanfang Liu
Summary: This study extracts individual daily activity-travel patterns from massive mobile phone network data, revealing the complex relationship of LAMs between workdays and day-offs, and identifying the formation mechanism of correlation patterns.
COMPUTERS ENVIRONMENT AND URBAN SYSTEMS
(2021)
Article
Public, Environmental & Occupational Health
Franziska Hommes, Achim Doerre, Susanne C. Behnke, Klaus Stark, Mirko Faber
Summary: The risk of giardiasis in travelers from Germany varies by destination, with Southern Asia being the most common exposure region. Latin America also has a significant number of cases, while Sub-Saharan Africa has a lower risk.
JOURNAL OF TRAVEL MEDICINE
(2023)
Review
Geochemistry & Geophysics
S. J. Arrowsmith, D. T. Trugman, J. MacCarthy, K. J. Bergen, D. Lumley, M. B. Magnani
Summary: Big Data Seismology, based on observations of inherently undersampled ground motion, is revolutionizing the field of seismology through new opportunities in earthquake processes understanding and resolving Earth structure. This review explores the development of new data-dense sensor systems, computing improvements, and new techniques and algorithms, and discusses the challenges and opportunities presented by Big Data Seismology. Recent scientific advances enabled by dense seismic data sets are also examined, highlighting the potential for significant progress in the field.
REVIEWS OF GEOPHYSICS
(2022)
Article
Economics
Caio Pieroni, Mariana Giannotti, Bianca B. Alves, Renato Arbex
Summary: This study analyzed the temporal and spatial patterns of urban transit movements in precarious settlement areas in Sao Paulo, Brazil using smart card data mining. The results revealed differences in travel behavior between low-income residents from precarious settlements and middle/high-income-class residents, with a focus on identifying low-paid employment travel patterns. The empirical evidence highlights smart card data's potential in uncovering low-paid employment spatial and temporal patterns.
JOURNAL OF TRANSPORT GEOGRAPHY
(2021)
Article
Computer Science, Information Systems
Masurah Mohamad, Ali Selamat, Ondrej Krejcar, Ruben Gonzalez Crespo, Enrique Herrera-Viedma, Hamido Fujita
Summary: This study proposes an alternate data extraction method, named CFS-DRSA, to enhance classifier performance and address challenges in large and problematic datasets. The method includes two crucial feature extraction tasks and has been experimentally validated to have high accuracy and credibility.
Article
Computer Science, Interdisciplinary Applications
Suyel Namasudra, S. Dhamodharavadhani, R. Rathipriya, Ruben Gonzalez Crespo, Nageswara Rao Moparthi
Summary: Big data is a combination of structured, semistructured, and unstructured data from various sources that needs to be processed before using it. Anomalies in big data refer to unusual occurrences of data that don't fit general patterns, which is a major problem. The Data Trust Method (DTM) is a technique that identifies and replaces untrustworthy data in big data using interpolation. This article discusses the application of DTM in improving the forecast quality of univariate time series (UTS) in big data using a neural network (NN) model.
Article
Green & Sustainable Science & Technology
Wei Yu, Jun Chen, Xingchen Yan
Article
Environmental Sciences
Wei Yu, Tao Wang, Yujie Xiao, Jun Chen, Xingchen Yan
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2020)
Article
Engineering, Electrical & Electronic
Wei Yu, Yan Zheng, Yongqiang Zhang
Summary: This research proposes a quantitative analysis idea for the carbon emission reduction of power battery echelon utilization. By selecting the appropriate battery type and capacity margin, and calculating the emissions at various stages in different life cycles, the reduction of carbon emission by the echelon utilization of retired power batteries is obtained. Finally, the overall carbon emissions that might be reduced in the future are calculated based on battery loading volume.
WORLD ELECTRIC VEHICLE JOURNAL
(2022)