nach oben

2019 | Buch

Kapitel lesen Erstes Kapitel lesen

Big Social Data and Urban Computing

First Workshop, BiDU 2018, Rio de Janeiro, Brazil, August 31, 2018, Revised Selected Papers

herausgegeben von: Jonice Oliveira, Claudio M. Farias, Dr. Esther Pacitti, Giancarlo Fortino

Verlag: Springer International Publishing

Buchreihe : Communications in Computer and Information Science

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten

Über dieses Buch

This book constitutes the thoroughly refereed proceedings of the First Big Social Data and Urban Computing Workshop, BiDU 2018, held in Rio de Janeiro, Brazil, in August 2018. The 11 full papers presented were carefully reviewed and selected from 40 submissions. The papers are organized in topical sections on urban mobility, urban sensing, contemporary social problems, collaboration and crowdsourcing.

Inhaltsverzeichnis

Frontmatter

Urban Mobility

Frontmatter

Characterizing Usage Patterns and Service Demand of a Two-Way Car-Sharing System

Abstract

Urban mobility is directly linked to the demand for communication resources and, clearly, its understanding is useful for better planning of urban and communication systems. However, getting data about urban mobility is still a challenge. In many cases, only a few companies have access to accurate and updated data. In most cases, these data are also privacy sensitive. It is thus important to generate models that can help to understand mobility patterns. We here characterize the demands of a two-way car-sharing system. We explore data of the public API of Modo, a car-sharing system that operates in Vancouver (Canada) and nearby regions. Our study uncovers patterns of users’ habits and demands in the service, which can be explored for urban and communication planning.

Felipe Rooke, Victor Aquiles, Alex Borges Vieira, Jussara M. Almeida, Idilio Drago

MobilityMirror: Bias-Adjusted Transportation Datasets

Abstract

We describe customized synthetic datasets for publishing mobility data. Companies are providing new transportation modalities, and their data is of high value for integrative transportation research, policy enforcement, and public accountability. However, these companies are disincentivized from sharing data not only to protect the privacy of individuals (drivers and/or passengers), but also to protect their own competitive advantage. Moreover, demographic biases arising from how the services are delivered may be amplified if released data is used in other contexts.

We describe a model and algorithm for releasing origin-destination histograms that removes selected biases in the data using causality-based methods. We compute the origin-destination histogram of the original dataset then adjust the counts to remove undesirable causal relationships that can lead to discrimination or violate contractual obligations with data owners. We evaluate the utility of the algorithm on real data from a dockless bike share program in Seattle and taxi data in New York, and show that these adjusted transportation datasets can retain utility while removing bias in the underlying data.

Luke Rodriguez, Babak Salimi, Haoyue Ping, Julia Stoyanovich, Bill Howe

MODAL - A Platform for Mobility Analyses Using Open Datasets

Abstract

Cities are becoming smart environments with the use of information and communication technologies (ICT). Data from these technologies are stored by various devices spread throughout the city and are available in open data portals, which can be used to improve essential services such as public transport and fed into platforms for visualization and analyses. Human and urban mobility analyses demonstrate that understanding movement patterns can assist governments in city’s decision-making process, as well as improve life quality of citizens. Aiming to enable mobility analysis in different cities, this work presents MODAL platform. This platform replicates mobility analyses and algorithms on databases of different cities using data obtained from open data portals. We assess the platform with a case study performing analyses of the transportation displacement within three different cities using complex network metrics. The results demonstrated the public transportation system efficiency showing regions of Chicago, Dubai and Taichung well served and regions which are key points to the transportation city interconnecting various areas. Moreover, we could evaluate how improved the transportation system would be by adding new lines or new transport system. The analyses demonstrated the platform potential to be used as support decision system for governments, showing the possibility of applying open data to improve city services and facilitate the conduction of analyses on various cities.

Wender Zacarias Xavier, Humberto Torres Marques-Neto

Urban Sensing

Frontmatter

Mensageria: A Smart City Framework for Real-Time Analysis of Traffic Data Streams

Abstract

Several smart city systems have focused on addressing a specific mobility problem scenario (e.g., air pollution, traffic jam) in a given city. The task of adding, extending, or porting the smart city scenario to other cities can be very challenging due to the rigid structure of such existing systems. To address this issue, in this paper we investigate common programming constructors that can be used to leverage the construction of such dynamic, smart city systems in the mobility domain. We propose Mensageria, a framework based on both the Complex Event Processing data-streaming processing paradigm and relational database management systems, which can dynamically deploy new or extend existing smart city scenarios in near real-time and maintain an updated dataset for provenance purposes. Mensageria provides several real-time primitives, such as filter, join, and enrich, that can be used to integrate, process, and analyze the city entities data streams. We discuss the generality, performance, and limitations of the proposed constructs through a real-world case study that was used in the Olympic Games of Rio in 2016 to detect, in real-time, existing and new situations that could affect the city mobility infrastructure.

Marcos Roriz Junior, Rafael Pereira de Oliveira, Felipe Carvalho, Sergio Lifschitz, Markus Endler

SLEDS: A DSL for Data-Centric Storage on Wireless Sensor Networks

Abstract

The dynamicity requirements of urban sensor networks rise new challenges to the development of data management and storage models. Software component techniques allow developers to build a software system from reusable, existing components sharing a common interface. Moreover, the development of urban sensor networks applications would greatly benefit from the existence of a dedicated programming environment. This paper proposes SLEDS, a Domain-Specific Language for Data-Centric Storage on Wireless Sensor Networks. The language includes high-level composition primitives, to promote a flexible coordination execution flow and interaction between components. We present the language specification as well as a case study of data storage coordination on sensor networks. The current specification of the language generates code for the NS2 simulation environment. The case study shows that the language implements a flexible model, which is general enough to be used on a wide variety of sensor network applications.

Marcos Aurélio Carrero, Martin A. Musicante, Aldri Luiz dos Santos, Carmem S. Hara

Extraction and Exploration of Business Categories Signatures

Abstract

Different business types may have distinct businesses functioning dynamics, i.e., popularity times, that can be dictated not only by the service offered but also due to other aspects. Performing the business popularity time comprehension allows us, for instance, to use this information as a business descriptor that could be explored in new services. Recently, Google launched a service, namely Popular Times, which provides the popularity times of commercial establishments. In this study, we collected and analyzed a large-scale dataset provided by that service for business in different cities in Brazil and in the United States. Our main contributions are: (1) clustering and analysis of the collected business popularity times dataset in each studied city; (2) approach for identifying the signature that represents the behavior of specific categories of venues; (3) training and evaluation of an inference model for categories of establishments; (4) user evaluation of some of our results.

Leonardo de Assis da Silva, Thiago H. Silva

Contemporary Social Problems

Frontmatter

Comparing Emotional Reactions to Terrorism Events on Twitter

Abstract

Over the last years, terrorism attempts have threatened the global population safety, impacting people in a complex emotional way. In this paper, we apply deep learning techniques to classify emotions of terrorism events, and develop a comparative analysis about emotional reactions on four events based on the demographics of tweeters, particularly gender, age and location. Our research questions involve comparing these events in terms of emotional shift, emotions according to age and gender, emotional reaction according to the closeness of the event and number/type of victims, as well as the terms used to express emotional reactions. The main conclusions were: fear, anger and sadness are the most expressed emotions; the emotions can be related to gender (e.g. fear for women, and anger for men); emotions seem to be not related to the closeness of the events, but seem to be affected by the casualties (number of kills/injuries); tweeters expressing fear and sadness tend to share words of affection and support, while tweeters expressing anger tend to use intense words of hate, intolerance and anger.

Jonathas G. D. Harb, Karin Becker

Using Government Data to Uncover Political Power and Influence of Contemporary Slavery Agents in Brazil

Abstract

This work uses open data published by the Brazilian government to investigate connections between agents involved on contemporary slavery labor and politicians, evaluating their power and influence. A network was built on data from Brazilian elections and campaign donations since 2002, including all candidates and donors associated to slave labor. Not only 263 direct candidatures from slavery agents were identified, but also more than 40 million Brazilian Reais in campaign donations for candidates for all electoral positions, showing a strong relation between slavery agents and Brazilian politicians. Data were also analyzed using metrics based on sociologist Manuel Castells’ Network Theory of Power that measure how much power and influence each donation is accounted for, in addition to its absolute amount. The resulting network was semantically enriched and modeled according to existing ontologies and published in RDF using Linked Open Data standards in a semantic knowledge graph, allowing information to be identified, disambiguated and interconnected by software agents in future research.

Letícia Dias Verona, Giseli Rabello Lopes, Maria Luiza Machado Campos

Collaboration and Crowdsourcing

Frontmatter

CidadeSocial: An Application Software for Opportunistic and Collaborative Engagement of Urban Populations

Abstract

The combination of mobile devices and easy access to the Internet enhances the spread of information coming from our day-to-day lives. People share all types of information such as events, opinions and problems in urban areas. Also, they can act as sensors by monitoring and sharing information about the demands made by inhabitants for urban changes on social media. Moreover, these platforms allow people to support each other - even strangers - through questions and answers, recommendations and indications. However, much of this kind of information is lost. Even social media (e.g. Facebook and Twitter) limit the spread of this flow of information up to the network borders, because their focus is on a relationship network. Consequently, the information does not reach people outside these networks. This paper describes an application software - named CidadeSocial – that allows inhabitants to share information according to their common interests. The application software exploits the geospatial location of the users to create a temporal social network, provides recommendations based on their profile and uses a gamification approach to encourage user engagement. Thus, we argue that CidadeSocial is a tool with potential to serve as an interface for engagement in improving the day-to-day life of cities and their inhabitants.

Ana Clara Correa, Eliel Roger, Tiago Cruz de França, José O. Gomes, Jonice Oliveira

Structures of Interactions and Data in Urban Networks: The Case of PortoAlegre.cc

Abstract

Urban spaces have been occupied by the massive use of new information and communication technology as digital social networks and platforms. The digital dimension of cities became a bidirectional and omnipresent path, creating relational and interactional structures able to exchange data and media. The networked city may be analyzed and debated as a complex system that demands research about communicational plurality and development of urban space representation, considering the increasing of informational and communicational density. Digital traces from social networks developed in PortoAlegre.cc, a collaborative web map registering issues and use of urban space, were used as data input for this research. Social network analysis was used as approach permitting network structures evaluation. The results reveal the existence of short paths, with predominance of structures that follow Small World model. The analyses showed efficient networks for data exchange increasing informational and communicational density. This work contributes for Urban computing bringing alternative approaches and perspectives for this multidisciplinary area with representation and knowledge that enhance the debate about urban space.

Pablo Vieira Florentino, Gilberto Corso Pereira

DMEK: Improving Profile Matching in Opportunistic Collaborations

Abstract

As the number of mobile devices grow, also grows the amount of data exchanged. This ever growing amount of data may overload Internet Service Providers. A possible solution to this problem is to use the mobile devices wireless network capabilities to exchange data by creating mobile P2P networks. These networks should opportunistically collaborate to exchange information to other devices in their proximity, only requiring users to specify their interests. This paper presents DMEK, (Decision Mobile Exchange of Knowledge) a solution where mobile devices disseminate knowledge among their users, opportunistically, using a decision mechanism based on profile matching. Experiments show DMEK feasibility and performance.

José Guilherme Mayworm, Jonice Oliveira, Fabrício Firmino, Claudio M. de Farias

Backmatter

Titel: Big Social Data and Urban Computing
herausgegeben von: Jonice Oliveira
Claudio M. Farias
Dr. Esther Pacitti
Giancarlo Fortino
Verlag: Springer International Publishing
Electronic ISBN: 978-3-030-11238-7
Print ISBN: 978-3-030-11237-0
DOI: https://doi.org/10.1007/978-3-030-11238-7